ANALYZING THE JOURNAL COVERAGE OF ABSTRACTING INDEXING DATABASES AT VARIABLE AGGREGATE AND ANALYTIC LEVELS/

Several studies have attempted to reveal the real scope, completeness, and consistency of journal coverage across time in abstracting/indexi ng publications. Online and CD-ROM databases make easier the collectio n of posting information for journals to determine the traits of their coverage but do not eliminate entirely the tedium of manual methods. Difficulties are compounded when multiple databases are used to compar e the journal coverage and overlap of several databases for an extende d time period. This article discusses the use of advanced search comma nds and the Journal Name Finder database of DIALOG, to simplify the co llection and processing of posting information for 42 prestigious libr ary and information science (LIS) serials between 1966 and 1996 in six databases. It presents the major types of deficiencies in journal cov erage that may yield very incomplete search results and may distort th e results of bibliometric and scientometric studies. The methodology a llows the creation of various subsets by the versatile combination of journal titles, time periods and databases, and encourages longitudina l data collection and comparison of variously aggregated datasets. It can be used for profiling the journal coverage in the evaluation of da tabases in support of bibliometric research, database acquisition, and licensing decisions.